piaoquan_crawler/piaopiaoquan/piaopiaoquan/piaopiaoquan_sift.py

# -*- coding: utf-8 -*-
# @Author: zhangyong
# @Time: 2023/11/30
import json
import os
import sys
import time
import uuid
from hashlib import md5

from appium import webdriver
from appium.webdriver.extensions.android.nativekey import AndroidKey
from bs4 import BeautifulSoup
from selenium.common.exceptions import NoSuchElementException
from selenium.webdriver.common.by import By
import multiprocessing


sys.path.append(os.getcwd())
from common import AliyunLogger, PiaoQuanPipeline, get_redirect_url
from common.common import Common
from common.mq import MQ


class PPQSiftRecommend:
    env = None
    driver = None
    log_type = None

    def __init__(self, log_type, crawler, env, rule_dict, our_uid):
        self.mq = None
        self.platform = "piaopiaoquan-sift"
        self.download_cnt = 0
        self.element_list = []
        self.count = 0
        self.swipe_count = 0
        self.log_type = log_type
        self.crawler = crawler
        self.env = env
        self.rule_dict = rule_dict
        self.our_uid = our_uid
        if self.env == "dev":
            chromedriverExecutable = "/Users/tzld/Downloads/chromedriver_v111/chromedriver"
        else:
            chromedriverExecutable = "/Users/tzld/Downloads/chromedriver_v111/chromedriver"

        Common.logger(self.log_type, self.crawler).info("启动微信")
        # 微信的配置文件
        caps = {
            "platformName": "Android",
            "devicesName": "Android",
            # "platformVersion": "11",
            # "udid": "emulator-5554",
            "appPackage": "com.tencent.mm",
            "appActivity": ".ui.LauncherUI",
            "autoGrantPermissions": "true",
            "noReset": True,
            "resetkeyboard": True,
            "unicodekeyboard": True,
            "showChromedriverLog": True,
            "printPageSourceOnFailure": True,
            "recreateChromeDriverSessions": True,
            "enableWebviewDetailsCollection": True,
            "setWebContentsDebuggingEnabled": True,
            "newCommandTimeout": 6000,
            "automationName": "UiAutomator2",
            "chromedriverExecutable": chromedriverExecutable,
            "chromeOptions": {"androidProcess": "com.tencent.mm:appbrand0"},
        }
        try:
            self.driver = webdriver.Remote("http://localhost:4723/wd/hub", caps)
        except Exception as e:
            print(e)
            AliyunLogger.logging(
                code="3002",
                platform=self.platform,
                mode=self.log_type,
                env=self.env,
                message=f'appium 启动异常: {e}'
            )
            return
        self.driver.implicitly_wait(30)
        for i in range(120):
            try:
                if self.driver.find_elements(By.ID, "com.tencent.mm:id/f2s"):
                    Common.logger(self.log_type, self.crawler).info("微信启动成功")
                    AliyunLogger.logging(
                        code="1000",
                        platform=self.platform,
                        mode=self.log_type,
                        env=self.env,
                        message="启动微信成功"
                    )
                    break
                else:
                    pass
            except NoSuchElementException:
                AliyunLogger.logging(
                    code="3001",
                    platform=self.platform,
                    mode=self.log_type,
                    env=self.env,
                    message="打开微信异常"
                )
                time.sleep(1)

        Common.logger(self.log_type, self.crawler).info("点击漂漂圈精选视频群聊")
        self.driver.find_elements(By.XPATH, '//*[@text="漂漂圈精选视频"]')[-1].click()

        time.sleep(5)
        Common.logger(self.log_type, self.crawler).info('点击"漂漂圈丨福年"卡片')
        self.driver.find_elements(By.XPATH, '//*[@text="漂漂圈丨福年"]')[-1].click()
        AliyunLogger.logging(
            code="1000",
            platform=self.platform,
            env=self.env,
            mode=self.log_type,
            message="打开漂漂圈丨福年卡片成功进入小程序"

        )

        time.sleep(5)
        self.driver.press_keycode(AndroidKey.BACK)
        AliyunLogger.logging(
            code="1000",
            platform=self.platform,
            env=self.env,
            mode=self.log_type,
            message="进入漂漂圈丨福年 更多热门成功"

        )
        time.sleep(5)

        self.get_videoList()
        time.sleep(1)
        self.driver.quit()

    def search_elements(self, xpath):
        time.sleep(1)
        windowHandles = self.driver.window_handles
        for handle in windowHandles:
            self.driver.switch_to.window(handle)
            time.sleep(1)
            try:
                elements = self.driver.find_elements(By.XPATH, xpath)
                if elements:
                    return elements
            except NoSuchElementException:
                pass

    def check_to_applet(self, xpath):
        time.sleep(1)
        webViews = self.driver.contexts
        self.driver.switch_to.context(webViews[-1])
        windowHandles = self.driver.window_handles
        for handle in windowHandles:
            self.driver.switch_to.window(handle)
            time.sleep(1)
            try:
                self.driver.find_element(By.XPATH, xpath)
                Common.logger(self.log_type, self.crawler).info("切换到WebView成功\n")
                AliyunLogger.logging(
                    code="1000",
                    platform=self.platform,
                    mode=self.log_type,
                    env=self.env,
                    message="成功切换到 webview"
                )
                return
            except NoSuchElementException:
                time.sleep(1)

    def swipe_up(self):
        self.search_elements('//*[@class="single--title"]')
        size = self.driver.get_window_size()
        self.driver.swipe(int(size["width"] * 0.5), int(size["height"] * 0.8),
                          int(size["width"] * 0.5), int(size["height"] * 0.442), 200)
        self.swipe_count += 1

    def get_video_url(self, video_title_element):
        for i in range(3):
            self.search_elements('//*[@class="single--title"]')
            Common.logger(self.log_type, self.crawler).info(f"video_title_element:{video_title_element[0]}")
            time.sleep(1)
            Common.logger(self.log_type, self.crawler).info("滑动标题至可见状态")
            self.driver.execute_script("arguments[0].scrollIntoView({block:'center',inline:'center'});",
                                       video_title_element[0])
            time.sleep(3)
            Common.logger(self.log_type, self.crawler).info("点击标题")
            video_title_element[0].click()
            self.check_to_applet(xpath=r'//wx-video[@class="infos--flex-box infos--title-wrap"]')
            Common.logger(self.log_type, self.crawler).info("点击标题完成")
            time.sleep(10)
            video_url_elements = self.search_elements(
                '//wx-video[@class="dynamic-index--video-item dynamic-index--video"]')
            if video_url_elements:
                return video_url_elements[0].get_attribute("src")

    def parse_detail(self, index):
        page_source = self.driver.page_source
        soup = BeautifulSoup(page_source, 'html.parser')
        soup.prettify()
        video_list = soup.findAll(name="wx-view", attrs={"class": "single--dynamic-item"})
        element_list = [i for i in video_list][index:]
        return element_list[0]

    def get_video_info_2(self, video_element):
        Common.logger(self.log_type, self.crawler).info(f"本轮已抓取{self.download_cnt}条视频\n")
        if self.download_cnt >= int(self.rule_dict.get("videos_cnt", {}).get("min", 10)):
            self.count = 0
            self.download_cnt = 0
            self.element_list = []
            return
        self.count += 1
        Common.logger(self.log_type, self.crawler).info(f"第{self.count}条视频")
        # 获取 trace_id, 并且把该 id 当做视频生命周期唯一索引
        trace_id = self.crawler + str(uuid.uuid1())
        AliyunLogger.logging(
            code="1001",
            platform=self.platform,
            mode=self.log_type,
            env=self.env,
            trace_id=trace_id,
            message="扫描到一条视频",
        )
        # 标题
        video_title = video_element.find("wx-view", class_="single--title").text
        # 播放量字符串
        play_str = video_element.find("wx-view", class_="single--favor-text").text
        user_name = video_element.find("wx-view", class_="single--nick").text
        # 头像 URL
        avatar_url = video_element.find("wx-image", class_="single--avatar-image")["src"]
        # 封面 URL
        cover_url = video_element.find("wx-image", class_="single--image")["src"]
        play_cnt = play_str.replace("+", "").replace("次播放", "")
        if "万" in play_cnt:
            play_cnt = int(play_cnt.split("万")[0]) * 10000
        out_video_id = md5(video_title.encode('utf8')).hexdigest()
        out_user_id = md5(user_name.encode('utf8')).hexdigest()

        video_dict = {
            "video_title": video_title,
            "video_id": out_video_id,
            'out_video_id': out_video_id,
            "duration_str": '',
            "duration": 0,
            "play_str": play_str,
            "play_cnt": play_cnt,
            "like_str": "",
            "like_cnt": 0,
            "comment_cnt": 0,
            "share_cnt": 0,
            "user_name": user_name,
            "user_id": out_user_id,
            'publish_time_stamp': int(time.time()),
            'publish_time_str': time.strftime("%Y-%m-%d %H:%M:%S", time.localtime(int(time.time()))),
            'update_time_stamp': int(time.time()),
            "avatar_url": avatar_url,
            "cover_url": cover_url,
            "session": f"piaopiaoquan_sift-{int(time.time())}"
        }
        pipeline = PiaoQuanPipeline(
            platform=self.crawler,
            mode=self.log_type,
            item=video_dict,
            rule_dict=self.rule_dict,
            env=self.env,
            trace_id=trace_id
        )
        flag = pipeline.process_item()
        if flag:
            video_title_element = self.search_elements(f'//*[contains(text(), "{video_title}")]')
            if video_title_element is None:
                return
            Common.logger(self.log_type, self.crawler).info("点击标题，进入视频详情页")
            AliyunLogger.logging(
                code="1000",
                platform=self.platform,
                mode=self.log_type,
                env=self.env,
                message="点击标题，进入视频详情页",
            )
            video_url = self.get_video_url(video_title_element)
            video_url = get_redirect_url(video_url)
            if video_url is None:
                self.driver.press_keycode(AndroidKey.BACK)
                time.sleep(5)
                return
            video_dict['video_url'] = video_url
            video_dict["platform"] = self.crawler
            video_dict["strategy"] = self.log_type
            video_dict["out_video_id"] = video_dict["video_id"]
            video_dict["crawler_rule"] = json.dumps(self.rule_dict)
            video_dict["user_id"] = self.our_uid
            video_dict["publish_time"] = video_dict["publish_time_str"]
            print(video_dict)
            self.driver.press_keycode(AndroidKey.BACK)
            self.mq.send_msg(video_dict)
            self.download_cnt += 1
            time.sleep(5)

    def get_video_info(self, video_element):
        try:
            self.get_video_info_2(video_element)
        except Exception as e:
            Common.logger(self.log_type, self.crawler).error(f"抓取单条视频异常:{e}\n")
            AliyunLogger.logging(
                code="3001",
                platform=self.platform,
                mode=self.log_type,
                env=self.env,
                message=f"抓取单条视频异常:{e}\n"
            )

    def get_videoList(self):
        self.mq = MQ(topic_name="topic_crawler_etl_" + self.env)
        self.driver.implicitly_wait(20)
        # 切换到 web_view
        self.check_to_applet(xpath='//*[@class="single--item-content"]')
        print("切换到 webview 成功")
        time.sleep(1)
        if self.search_elements('//*[@class="single--item-content"]') is None:
            Common.logger(self.log_type, self.crawler).info("窗口已销毁\n")
            AliyunLogger.logging(
                code="3000",
                platform=self.platform,
                mode=self.log_type,
                env=self.env,
                message="窗口已销毁"
            )
            self.count = 0
            self.download_cnt = 0
            self.element_list = []
            return

        print("开始获取视频信息")
        for i in range(50):
            print("下滑{}次".format(i))
            element = self.parse_detail(i)
            self.get_video_info(element)
            self.swipe_up()
            time.sleep(1)
            if self.swipe_count > 100:
                return

        print("下滑完成")
        Common.logger(self.log_type, self.crawler).info("已抓取完一组，休眠 5 秒\n")
        AliyunLogger.logging(
            code="1000",
            platform=self.platform,
            mode=self.log_type,
            env=self.env,
            message="已抓取完一组，休眠 5 秒\n",
        )
        time.sleep(5)


def run():
    rule_dict1 = {"period": {"min": 365, "max": 365},
                  "duration": {"min": 0, "max": 0},
                  "favorite_cnt": {"min": 0, "max": 0},
                  "videos_cnt": {"min": 5000, "max": 0},
                  "share_cnt": {"min": 0, "max": 0}}
    PPQSiftRecommend("recommend", "piaopiaoquan", "dev", rule_dict1, [64120158])


if __name__ == "__main__":
    process = multiprocessing.Process(
        target=run
    )
    process.start()
    while True:
        if not process.is_alive():
            print("正在重启")
            process.terminate()
            time.sleep(60)
            os.system("adb forward --remove-all")
            process = multiprocessing.Process(target=run)
            process.start()
        time.sleep(60)